FormalPara Key Summary Points

Why carry out this study?

Though oral antiviral medications are important tools to prevent severe COVID-19 outcomes, uptake remains low for reasons that are not entirely understood.

We investigated the relationship between perceived risk for severe COVID-19 outcomes and use of oral antiviral treatment for COVID-19 among those who were eligible for treatment.

What was learned from the study?

Out of 934 non-institutionalized US adults surveyed in April 2023 who were antiviral-eligible with at least one confirmed SARS-CoV-2 infection since December 1, 2021, only 18.5% considered themselves to be at high risk for severe COVID-19 and 15.9% took oral antivirals within 5 days of SARS-CoV-2 infection.

Perceived high-risk status was associated with being more likely to be aware, to be prescribed, and to take oral antivirals at any time or within 5 days of infection.

Differences between perceived and actual risk for severe COVID-19 may partially explain low oral antiviral uptake.

Introduction

In addition to COVID-19 vaccination [1], oral antiviral medications such as nirmatrelvir/ritonavir and molnupiravir are important tools for preventing severe COVID-19 outcomes. These drugs were first approved by the Federal Drug Administration (FDA) for Emergency Use Authorization (EUA) in December 2021 for individuals at high risk for severe COVID-19 disease [2, 3], based on clinical trials in the summer and fall of 2021 showing substantial reductions in hospitalizations and deaths among high-risk, non-hospitalized and unvaccinated adults with SARS-CoV-2 infection [4, 5]. In the Omicron era, real-world effectiveness of nirmatrelvir/ritonavir against COVID-19-related hospitalizations, disease progression, and death has been demonstrated in a number of different outpatient settings, including among individuals with prior SARS-CoV-2 immunity through infection or vaccination, with a high level of protection for those older than 65 or with comorbidities [6,7,8,9,10,11,12,13]. Evidence of molnupiravir effectiveness among high-risk vaccinated adults, however, is mixed [14,15,16,17,45], underscoring the importance of physician recommendation for improving oral antiviral uptake.

Our findings also give some insight into other possible reasons for non-use; we found that among eligible individuals who were not prescribed oral antivirals for their last SARS-CoV-2 infection, more than half did not think they needed it, while a healthcare provider refused a prescription for less than 5% of individuals. Using data from three rounds of the US Census Household Pulse Survey collected between June and August 2022, an analysis of reasons for not receiving oral antiviral treatment among individuals 65 or older with at least mild symptoms cited not feeling very sick (20.6%) and not thinking they needed treatment (16.4%) as the most common reasons for not taking oral antivirals, followed by not having the treatment offered or recommended to them by a healthcare provider (14.5%) [46]. The same study also found that not seeing oneself as a member of a high-risk group was a fairly uncommon reason cited for not taking oral antivirals (4.9%). This is surprising, given our finding that the majority of antiviral-eligible individuals do not perceive themselves to be at high risk for severe COVID-19 outcomes.

Though the survey design used in this study used sampling and weighting methods to reduce any bias resulting from sociodemographic differences in recruitment and response rates and has been shown to provide reliable and nationally representative estimates for other COVID indicators [33], we were unable to account for potential nonresponse bias related to SARS-CoV-2 infection history or oral antiviral use. For example, if people who used oral antivirals were more likely to complete the survey, estimates may be inflated. However, our approach benefits from the ability to reflect perceived risk, COVID-19 outcomes, antiviral eligibility, and oral antiviral use among those who may not have accessed the healthcare system when they last had COVID-19.

Our study has several other limitations worth noting. Data were self-reported, and there is the potential for misclassification and reporting bias. However, studies comparing self-reported COVID-19 vaccine status with electronic health records or vaccine registries have found more than 90% agreement between the two [47,48,49]. Further, self-reported health conditions [50] and use of medication [51] have been found to have high specificity and moderate sensitivity; thus, we may have underestimated the prevalence of health conditions or oral antiviral use. Relatedly, our estimates of SARS-CoV-2 infection timing and oral antiviral use were likely more accurate for more recent infections. In addition, because our definition of SARS-CoV-2 infection was based on self-reported diagnosis by a health care provider or testing, we would have missed asymptomatic infections and symptomatic infections among those who did not receive a COVID test or seek care (i.e., possible infections), leading to an underestimated prevalence of those with a history of SARS-CoV-2 infection. In our sample, none of those with possible infections reported antiviral prescription. This group requires further investigation, however, because they may have barriers to knowledge, access to testing, and access to care that could get in the way of antiviral use.

The cross-sectional nature of this study made it impossible to determine the temporality of responses, since risk and outcomes were assessed at the same time. While one might predict an individual’s perceived risk to influence their decisions around treatment once sick, one might also predict that experience with an illness might influence that individual’s perception of their risk going forward (e.g., if their symptoms were very mild). We did not ask about the severity or nature of symptoms experienced by individuals at the time of their last SARS-CoV-2 infection, which might be important factors to investigate further to better understand what influences an individual to seek antiviral treatment. In addition, participants were given only two possible reasons for non-use of oral antivirals, which limited our ability to better understand non-use. We plan to conduct a more granular evaluation of barriers to oral antiviral use, including elements of both patient and provider decision-making following the transition of nirmatrelvir/ritonavir and molnupiravir to the commercial market in November 2023 in the United States [52].

Conclusions

In conclusion, we found that less than 20% of antiviral-eligible adults (based on CDC criteria for high risk) perceived themselves to be at high risk for severe COVID-19 outcomes. This misperception of risk status, and thus antiviral eligibility, may be a contributor to suboptimal COVID-19 oral antiviral uptake to date, despite broad awareness of COVID-19 treatment options. Our findings suggest the need for additional strategies to ensure that individuals better under their COVID-19 risk, which may improve uptake of available treatment options.